The use of nucleic acid probe tests based on hybridization is hindered by a lack of sensitivity. Nucleic acid probe technology has been discovered for its value in the detection of various diseases, organisms or genetic defects, and thus, has developed quite rapidly. Current amplification techniques provide the sensitivity lacking in earlier assays and necessary to qualitatively determine the presence of minute quantities of nucleic acid. That is, they are sensitive enough to be able to detect a single copy of a specific nucleic acid sequence.
A number of methodologies are currently used for the measurement of gene expression. The most sensitive of these methodologies utilizes the polymerase chain reaction (PCR) technique, the details of which are provided in U.S. Pat. No. 4,683,195, U.S. Pat. No. 4,683,202, and U.S. Pat. No. 4,965,188, all to Mullis et al., all of which are specifically incorporated herein by reference. The details of PCR technology, thus, are not included herein. Recently, additional technologies for the amplification of nucleic acids have been described, most of which are based upon isothermal amplification strategies as opposed to the temperature cycling required for PCR. These strategies include, for example, Strand Displacement Amplification (SDA)(U.S. Pat. Nos. 5,455,166 and 5,457,027 both to Walker; Walker et al. (1992) PNAS 89:392; each of which is specifically incorporated herein by reference) and Nucleic Acid Sequence-Based Amplification (NASBA)(U.S. Pat. No. 5,130,238 to Malek et al.; European Patent 525882 to Kievits et al.; both specifically incorporated herein by reference). Each of these amplification technologies are similar in that they employ the use of short, deoxyribonucleic acid primers to define the region of amplification, regardless of the enzymes or specific conditions used. This invention is concerned primarily with a novel amplification-primer methodology useful for the amplification and subsequent quantitation of nucleic acid using any amplification method, including but not limited to the aforementioned strategies. In particular, this application describes a method for the quantitation of RNA transcripts using genomic DNA as an amplification competitor.
Generally, amplification of DNA molecules utilizes a thermostable polymerase, for example, Taq polymerase, described in European Patent 258017, specifically incorporated herein by reference. Until recently, RNA amplification required a separate, additional step and the use of non-thermostable reverse transcriptase enzymes to generate a cDNA capable of being amplified by a thermostable DNA polymerase, such as Taq. The discovery of a recombinant thermostable enzyme (rTth) capable of coupling reverse transcription of the RNA with DNA amplification in a single enzyme:single reaction procedure greatly simplified and enhanced RNA amplification (see, Myers & Gelfand (1991) Biochemistry 30:7661-7666; U.S. Pat. No. 5,407,800 to Gelfand and Myers, both specifically incorporated herein by reference).
Recombinant thermostable enzyme polymerase chain reaction (rTthPCR) is exquisitely sensitive. However, some notable difficulties are encountered when applying PCR to the quantitative measurement of gene expression. All PCR-based methods for measuring gene expression have certain features in common: 1) the nucleic acids from a chosen sample must be extracted; and 2) the target sequence is amplified by PCR, allowing detection and quantitation of that sequence. The efficiencies of both the extraction step and the amplification step directly affect the outcome of the assay. Further, the detection step reveals how many copies of the target sequence have been made, but not how many copies were contained in the original sample. To determine the number of original copies, the efficiency of the nucleic acid extraction, as well as the efficiency of each PCR reaction must be known. An absolute quantitation of sample copy number is difficult using the current state of extraction technologies--if not impossible. The best that can be achieved on a routine basis is quantifying changes in gene expression rate, rather than quantifying the exact numbers of copies of the target sequence contained in the sample. However, this quantification of gene expression rate changes can only be done if the efficiencies of the extraction and amplification processes are constant or can be controlled for. There are two basic types of control: the first is commonly known as exogenous control (Gilliland et al. (1990) PCR Protocols, Innis et al. ed., pp. 60-69, Academic Press; Wang et al. (1989) Proc. Natl. Acad. Sci. USA 86:9717-9721, both specifically incorporated herein by reference), and the second, endogenous control (Dveksler et al. (1992) PCR Methods and Applications 6:283-285; Spanakis (1993) Nucleic Acids Research 21:3809-3819, both specifically incorporated herein by reference).
Exogenous control involves the use of an artificially introduced DNA or RNA molecule that is added, either to the extraction step or to the PCR step, in a known concentration. The concept of adding an exogenous nucleic acid at a known concentration in order to act as an internal standard for quantitation was introduced by Chelly et al. (1988) Nature 333:858-860, specifically incorporated herein by reference. This approach was improved following the realization that PCR amplification efficiency is dependent upon the primer sequences (WO 91/02817 to Wang et al., specifically incorporated herein by reference). Therefore, utilizing a control fragment that is amplified with the same primers as the target sequence more accurately reflects target sequence amplification efficiency relative to the internal standard (see, for example, WO 93/02215 to Fox and Griffiths; WO 92/11273 to McCallum et al.; U.S. Pat. No. 5,213,961 to Bunn et al.; and, U.S. Pat. No. 5,219,727 to Wang et al., all of which are specifically incorporated herein by reference). Similar strategies have proven effective for quantitative measurement of nucleic acids utilizing isothermal amplification reactions such as NASBA (Kievits et al., supra) or SDA (Walker, supra).
The use of an endogenous control regulates variations in extraction efficiency. Control choice is important in that several requirements must be met in order for it to work. The first requirement is that the copy number of the control must remain constant; the second, the control must amplify with similar efficiency to the sequence being monitored. Several constitutively expressed genes have been considered as control candidates, since the expression of these genes is relatively constant over a variety of conditions. The most common of these are the .beta.-actin gene, the glyceraldehyde-3-phosphate dehydrogenase gene (GAPDH), and the 16S ribosomal RNA gene. While these genes are considered to be constitutively expressed, they are not ideal for the above-discussed purpose. The first problem is that under some conditions the expression level of these genes varies (Spanakis (1993) supra; Dveksler et al. (1992) supra) possibly invalidating them as controls. Further, these genes necessitate using a different set of PCR primers from any other gene being assayed, possibly leading to differences in PCR efficiency, and thus, invalidating them as controls. Zenilman et al. ((1995) Analytical Biochemistry 224:339-346, specifically incorporated herein by reference) attempted to design a competitive polymerase chain reaction method that altered the cDNA generated during reverse transcription using a composite primer to introduce a common intron primer sequence into the cDNA, such that both genomic DNA and cDNA could then be amplified with the same primer set. That method requires careful design of the composite primers so that, following reverse transcription, the melting temperature (Tm) differential between the composite and DNA primers is adjusted to prevent the composite primer from binding to the genomic DNA. It further requires separate steps for reverse transcription and DNA amplification, and uses a cumbersome, impractical radioactive detection methodology.
The detection of amplified products following quantitative PCR must provide a method of distinguishing the added control standard from the target nucleic acid sequence. Endogenous controls include, for example, radioactive probes specific to the control sequence, wherein the control and target detection is split into separate detection reactions for each amplification tube. Exogenous standards have been designed so as to be distinguishable by size of the amplimer as visualized on an agarose gel (Scadden et al. (1992) J. Infect. Dis. 165:1119-1123; Piatak et al. (1993) Biotechniques 14:70-80, both specifically incorporated herein by reference) or by introducing an internal restriction site through mutagenesis, wherein the restriction fragments are again detected on an agarose gel (Becker-Andre and Hahlbrock (1989) Nucleic Acid Res. 17:9437-9446; Steiger et al. (1991) J. Virol. Methods 34:149-160, both specifically incorporated herein by reference). All of these methods employ a relatively insensitive and imprecise detection method. More recently developed methods utilize chemiluminescent or enzyme-linked immunosorbent assay (ELISA) based detection that more precisely measure amplified product (Mulder et al. (1994) J. Clin. Microbiol. 32:292-300, specifically incorporated herein by reference).
It is preferable to utilize detection methods involving a homogeneous assay, wherein detection occurs concomitantly with amplification. This is possible utilizing the nuclease assay methodology described in U.S. Pat. No. 5,210,015 to Gelfand et al., specifically incorporated herein by reference. The preferred probes for the nuclease assay, those probes having the greatest specificity, are described by Marmaro et al. (U.S. Ser. No. 08/340,558; U.S. Ser. No. 08/559,405; and U.S. Ser. No. 08/558,303, the specifications of which are specifically incorporated herein by reference). These preferred probes, designated "TaqMan" probes and developed by the Perkin Elmer Corp. (Foster City, Calif.), utilize energy transfer fluorescence methods, wherein the probes are self-quenched until cleaved during the PCR nuclease assay. Cleavage during PCR generates an increased fluorescence signal. Furthermore, several reporter dyes can be monitored in the same reaction.
The characteristics of an ideal quantitation control include: its presence in unchanging concentrations in the cell; it possesses the same PCR efficiency as the system being assayed; and, it is easily distinguishable from the system being assayed. It is further desirable that the control is a natural component of the biological specimen making it unnecessary to synthesize and externally add an artificial standard, and the control must accurately accommodate fluctuations both in nucleic acid extraction as well as amplification efficiencies. Finally, the control must be accurately distinguishable from the target, preferably by an homogeneous assay of such specificity so as to have no cross reactivity that alters the ratio of the two. The quantitation methodology disclosed herein meets these requirements by integrating existing technologies with novel amplification and primer design techniques, as disclosed herein.